1 Introduction

Faced with increasing global energy demands, the development of renewable energies has become a crucial necessity to tackle the depletion of conventional fossil fuels and combat the rise in greenhouse gas emissions responsible for climate change [1]. Several countries have made significant commitments to reduce reliance on fossil fuels in electricity production and increase the use of clean energy. For example, Germany has pledged to achieve near energy independence through renewable energy sources by 2050 [2]. Similarly, in 2012, the Saudi Arabian government set an ambitious target to cover 30% of the country's electricity demand by 2032 by investing $109 billion in the renewable energy sector [3]. One of the proposed solutions to address the challenge of electricity scarcity for consumers is wind energy, which has emerged as one of the most promising and attractive sources of renewable energy globally. Its widespread availability and reduced environmental impact [4] make it a viable solution to meet the growing electricity needs in many regions around the world. According to the International Energy Agency (IEA), the installed global wind capacity reached 743 GW in 2020, representing a 19% increase compared to the previous year [5].

Wind energy is emerging as a crucial renewable energy source for electricity production in Africa. Several studies have been conducted to assess its potential, mainly focusing on average wind speed and important statistical parameters like annual or monthly Weibull parameters [6]. Among these studies exploring wind potential in different regions, the northwest of Africa has been the subject of research to evaluate wind potential in countries such as Mauritania, using data collected over multiple years [7]. Senegal, a country located in West Africa, has adopted wind energy to diversify its energy mix and reduce its dependence on imported fossil fuels [8]. Preliminary studies have been conducted to assess the wind potential of various sites along the northwestern coast of Senegal, including Gandon, Kayar, Mboro, Taiba Ndiaye, and Potou, using real data collected over a year [9,10,11]. As a result, the Taïba Ndiaye site, situated in the Thiès region, was chosen as the ideal location for the country’s first wind farm, expected to contribute approximately 15% to the national electricity production [12].

Furthermore, several studies emphasize the importance of meteorological conditions for wind energy production. Research conducted by Delft University of Technology shows that wind energy production is highly dependent on wind conditions, which vary significantly based on the region and season [13]. Similarly, a study by the University of Lisbon reveals that the amount of energy produced by a wind turbine depends on the wind speed it is exposed to [14]. In Senegal, preliminary studies conducted at Taïba Ndiaye before the installation of the wind farm clearly demonstrate that wind variations will have a significant impact on the future wind farm’s production [15]. Therefore, it is essential to consider these specific wind conditions to improve the prediction of wind energy production [16,17,18,19,20].

Therefore, it is important to understand how wind conditions influence wind energy production to improve the performance of wind farms. The Taïba Ndiaye region has been the subject of several preliminary studies. For example, [21] found that the annual average wind speed in Taïba Ndiaye was 6.63 m/s, with a maximum speed of 8.2 m/s. They also estimated that the installed capacity of the large-scale wind farm should be around 500 MW. The installed capacity of the Ndiaye wind farm is reported to be 620 GWh, with a power density of 293 W/m2. They even recommend the use of medium-sized turbines, with a hub height of 80 m and a rated power of 2.5 MW [22, 23] found that the energy production of the Taïba Ndiaye wind farm was 441 GWh during the studied year. They also found that energy production could be higher during the dry season than during the rainy season, with maximum production in December. However, few studies have been conducted on the temporal dynamics of production from this new wind farm.

This article analyses the dynamics of wind power production at the Taïba Ndiaye wind farm and its link to meteorological parameters at different time scales. Production data were evaluated on an annual, monthly, daily, hourly, and intra-day basis to better understand the factors that influence wind power production in this region of Senegal. Production data were also compared to wind conditions at 100 m above ground level to assess the influence of wind power production. This analysis could help improve and predict the performance of the wind farm, promote the development of a sustainable and diversified energy mix in Africa, and reduce greenhouse gas emissions. Moreover, it could serve as a benchmark for wind farms, enabling them to optimize and predict their performances more effectively.

The rest of the paper is organized as follows: the next section describes the Taïba Ndiaye wind farm, and the data used. The main results and discussions are provided in the following section, while the summary and conclusions are presented in the last section.

2 Data and methods

2.1 Presentation of the wind farm and data

The Taïba Ndiaye wind farm is located in a coastal area approximately 85 km northeast of Dakar, Senegal, where the winds are strong and consistent. This allows for maximizing the efficiency of the wind farm and producing a significant amount of renewable electricity. Developed by Lekela Power, a company specialized in renewable energy, the wind farm started producing electricity in 2020 to contribute to the country's goal of reaching a 30% share of renewable energy in electricity production by 2025 [24]. It is considered one of the largest wind farms in West Africa. The wind farm is equipped with 46 wind turbines, each with a capacity of 3.45 MW, for a total capacity of 158.7 MW which represents about 15% of the electricity production capacity of Senegal [25].

Production data for the Taïba Ndiaye wind farm were provided by Lekela Power and include annual, monthly, daily, and hourly data for the year 2021. Wind data at a height of 100 m were obtained from the ERA5 dataset, which is produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) [26]. ERA5 data is based on weather models and has a spatial resolution of 0.25° (approximately 31 km) and an hourly temporal resolution. The quality of ERA5 data has been validated by several scientific studies [27], which have shown that these data are accurate and reliable for wind condition analysis in different regions of the world.

2.2 Methods

This study aimed to analyze the energy production data collected from the Taïba Ndiaye wind farm in Senegal. The primary focus was on the energy production (measured in MWh) of the 46 wind turbines. The adopted approach enabled us to obtain detailed data at a high temporal resolution of approximately 10 min.

In consideration of the wind farm's mast height set at 117 m, ERA5 reanalysis data from the Copernicus Climate Change Service (C3S) Climate Data Store (CDS) [28, 29] were utilized for acquiring wind information. The ERA5 dataset provide essential parameters, namely, the latitudinal (u) and meridional (v) components of the wind, which were employed to calculate the wind speed (W). Subsequently, using the derived components, the wind intensity at 100 m (W100) was computed following the Eq. 1 [30, 31]:

$${\text{W}}_{100} = \sqrt {{\text{u}}^{2} + {\text{v}}^{2} }$$
(1)

and the wind direction θ is given by Eq. 2 [30, 31]:

$${\uptheta } = \frac{180}{{\uppi }}{\text{ mod}}\left( {{\uppi } + {\text{atan}}2\left( {{\text{v}},{\text{ u}}} \right),{ }2{\uppi }} \right)$$
(2)

  • where θ is considered to be 0° for a wind coming from the north, and the angle θ increases clockwise.

  • the atan2 function is an enhanced version of the standard arctangent function, which considers both components (u and v) of the wind to calculate the angle θ. As a result, atan2(u, v) accurately determines the wind direction, regardless of the wind vector's position in the quadrants. Setting θ to 0° for a wind coming from the north, the angle θ increases clockwise.

In conclusion, θ is the angle that indicates the direction of the wind with respect to the north. It ranges from 0° (when the wind comes from the north), to 90° (east wind), 180° (south wind), and 270° (west wind), all the way to 360° (when the wind comes from the north again). This angle measurement helps us understand the wind's orientation and where it is coming from in relation to the north direction.

3 Results and discussions

3.1 Representation of the diurnal cycle at different temporal scales

3.1.1 Monthly variability of the diurnal cycle of production

Figure 1 represents the monthly variability of the diurnal cycle of wind power production from the Taïba Ndiaye wind farm. Firstly, this graph shows peaks in autumn and winter, with a maximum production of 2040 MWh in January. Similarly, the lowest levels are recorded in summer, reaching approximately 430 MWh in September.

Fig. 1
figure 1

Monthly variability of the diurnal cycle of wind energy production

Full size image

Furthermore, a marked diurnal cycle is observed, particularly between November and April. Maximums are recorded during the night, between 8 p.m. and 8 a.m., while minimums occur between 9 a.m. and 7 p.m. From May onwards, there is a considerable drop in production and a reversal of the diurnal cycle trend, which evolves towards a bimodal pattern until August.

3.1.2 Seasonal variability of the production diurnal cycle

The hourly production for winter (December–February), spring (March–May), summer (June–August), and autumn (September–November) is represented in the Fig. 2 to observe seasonal trends of the diurnal cycle. As previously mentioned, the winter season records the production peaks (75.50 MWh), while the rainy season is characterized by minimal production levels, with an average of around 21 MWh. The daily variation shows a clear monomodal trend in winter, spring, and autumn, characterized by night-time production peaks (7 p.m.–8 a.m.) and daytime minimum levels (9 a.m.–6 p.m.). Conversely, during the rainy season, the trend is bimodal with two production peaks, one at 5 a.m. and the other at 6 p.m., separated by a minimum at 8 a.m.

Fig. 2
figure 2

Seasonal variability of the production diurnal cycle

Full size image

3.1.3 Average diurnal cycle of production at Taïba Ndiaye

In summary, the Fig. 3 illustrates the global diurnal cycle of the Taïba Ndiaye wind farm power production. On average, the production of the farm is 65.11 MW per hour every day. Night-time peaks exceed 80 MW/h between midnight and 8 a.m., while daytime lows are around 40 MW/h between 10 a.m. and 6 p.m. The lowest hourly production is recorded around 2 p.m., with a value of approximately 30 MW.

Fig. 3
figure 3

Global diurnal cycle of the Taïba Ndiaye wind farm power production

Full size image

3.2 Monthly and seasonal distribution of production

Figure 4 shows the monthly variability of the production of the Taïba Ndiaye wind farm in 2021.

Fig. 4
figure 4

Monthly variability of the production (in MWh) of the Taïba Ndiaye wind farm

Full size image

There is a very pronounced trend with maximums between December and May, and minimums between July and November. The peak of production is recorded in January, while the smallest value is noted in September.

To further substantiate our findings, we present in the Fig. 5, the seasonal distribution of electricity production from the Taïba Ndiaye wind farm, categorized into winter (December–February), spring (March–May), summer (June–August), and autumn (September–November).

Fig. 5
figure 5

Seasonal variability of the production (in MWh) of the Taïba Ndiaye wind farm in: winter (December–February), spring (March–May), summer (June–August), and autumn (September–November)

Full size image

The results confirm those previously reported in Fig. 4, with production peaking during winter and spring, accounting for 40.4% and 33% of the annual production, respectively. Conversely, minimal quantities are recorded during the rainy season, in summer, representing approximately 11.2% of the annual production. Production during the autumn is also low, with a yield of nearly 15.5%. Indeed, the rainy season may lead to weaker winds in certain regions of the Sahel due to modifications in atmospheric conditions [32,33,34].

3.3 Effects of 100 m’ wind on the production of the wind farm

3.3.1 Monthly effects

The present figure (Fig. 6) illustrates the monthly distribution of the diurnal cycle of power production from the power farm (in blue) and the 100 m wind speed (in red) at Taïba Ndiaye. The results clearly show that it is the wind speed (at a height of 100 m) that modulates the power production of the farm. Indeed, both indicators exhibit a positive correlation throughout the year. The month of January corresponds to the maximum values of production (2040 MWh) and wind speed (7.7 m/s), while the month of August corresponds to the minimum values of wind speed (~ 4.5 m/s) and production. It is also observed that the diurnal cycle of production is directly linked to that of the wind, with maxima during the night (9 m/s) and minima during the day (< 6 m/s). Furthermore, the bimodal behavior of production in August can also be attributed to that of the wind speed.

Fig. 6
figure 6

Monthly distribution of the diurnal cycle of power production from the power farm (in blue) and 100 m wind speed (in red) at Taïba Ndiaye

Full size image

The figure below (Fig. 7) shows the monthly wind roses at the Taïba Ndiaye site. It can be observed that the optimal wind directions for wind turbine production are recorded between the months of November and May, which correspond to the periods of maximum production. Indeed, the optimal direction for a wind turbine is generally between 30° and 60° from its axis, with an angular range where the wind blows regularly and consistently, corresponding to a wind direction ranging from northwest to northeast through north [35,36,37,38,39,40].

Fig. 7
figure 7

Monthly wind roses at the Taïba Ndiaye area

Full size image

The Fig. 7 clearly shows that during the months of maximum production (November to May), the wind direction is within the optimal intervals ranging from northwest to northeast. Starting from June, the wind direction gradually evolves from northwest to west, reaching its most western point around August, which also corresponds to the period of least energy production by the power farm.

3.3.2 Seasonal effects

The Fig. 8 illustrates the comparison between the seasonal distribution of the diurnal cycle of energy production and wind speed. As with the monthly data, it can be observed that the energy production of the Taïba Ndiaye wind farm depends on wind conditions. Indeed, there is a correlation between wind speed and energy production. The peaks in wind speed are observed during winter and spring, which are also the periods of maximum production. During these two seasons, wind speeds exceed 7 m/s, allowing for the generation of more than 63 MWh. Low winds in summer and autumn result in minimal energy production levels. Furthermore, seasonal production follows the diurnal variations of seasonal winds. For autumn, winter, and spring, the trend is unimodal, while for the rainy season in summer, it is bimodal. Finally, the diurnal cycle of these indicators reveals that production levels reach their maximum during the night due to strong winds that blow at that time of day.

Fig. 8
figure 8

Comparison between the seasonal distribution of the diurnal cycle of energy production and 100 m wind speed

Full size image

The analysis of seasonal wind directions in the Fig. 9 shows consistent trends. Optimal wind directions are observed in winter and spring, periods of maximum production for the wind power farm, when winds blow between the northwest and northeast. On the other hand, during summer and fall (autumn), periods of minimum production, wind directions shift to oscillate between northwest and west.

Fig. 9
figure 9

Seasonal wind directions at Taïba Ndiaye wind farm

Full size image

3.3.3 Analysis of night-time and daytime productions

The Fig. 10 presents the monthly production of our wind power farm for time intervals representing daytime (8 a.m.–6 p.m.) and night-time (7 p.m.–7 a.m.). We can observe that the production during the daytime and night-time follows a similar trend, reaching peaks between December and April and minima between June and November. However, the night-time production is always higher than that of the day, with a more pronounced difference between December and April. Starting in May, the farm produces approximately the same amount of electricity during the day and night. Overall, the wind farm’s production is 43.1% higher at night than during the day.

Fig. 10
figure 10

Monthly production of the wind power farm for time intervals representing day (8 a.m.–6 p.m.) and night (7 p.m.–7 a.m.)

Full size image

The graph presented below (Fig. 11) shows a seasonal comparison of power production between daytime and night-time. It is notable that the wind power farm produces more energy during the night than during the day, particularly during the winter months (December to February) and spring (March to May). For example, during the winter, night-time production is twice as high as daytime production. However, during the summer (the rainy season), energy production is approximately equal between day and night.

Fig. 11
figure 11

Seasonal comparison of electricity production between daytime and night-time

Full size image

The wind rose analysis confirms (in Fig. 12) that the winds are always located in the optimal directions mainly between northeast and northwest, both during the day and at night. However, it has been observed that during the night when the winds are stronger, the preferred direction is often north, while during the day, the winds blow more between north and northwest.

Fig. 12
figure 12

Global wind roses for daytime (left) and night-time (right)

Full size image

4 Conclusion

The aim of this article was to study the temporal dynamics of wind energy production at Taïba Ndiaye wind farm in Senegal using a multi-scale approach. Firstly, the monthly and seasonal distribution of production showed a marked trend, with peaks recorded between December and May (winter and spring), and minimums between July and November (summer and autumn). The diurnal cycle representation showed a pronounced variation, particularly between autumn and winter, with peaks during the night and minima during the day. In fact, night-time production was higher than daytime production by over 43%. Comparison of the 100 m wind speed and the farm production showed a strong positive correlation throughout the year. The maxima in production observed in winter and spring were caused by strong winds, while the lowest levels recorded during the summer season were attributable to the meteorological conditions. The wind direction indicates that the wind mainly blows in the optimal directions, especially in winter and spring, the periods of maximum production for the farm, when the winds are from the northwest to northeast. The results of this study could help better understand the behaviours of wind energy production and optimize the use of this renewable energy source in the region.